UCLouvain Economics Seminar

Can social media promote female access to political positions? Internet and social media have facilitated a number of evolutions regarding ideas, perceptions and attitudes towards women and their role in society. This paper focuses on the sub-Saharan African context, where female political under-representation is particularly salient, and which experienced a rising penetration of Facebook over the past decade. We build a novel panel dataset at the constituency-election level and exploit variation in the electoral outcomes observed across 8,162 races occurring in the context of 63 parliamentary elections in 17 countries. We leverage the staggered introduction of Facebook's Free Basics -i.e. free access to Facebook through partner mobile operators- across constituencies and time, and document the success of this connectivity shock and its subsequent effect on female political representation. We find that Facebook's Free Basics significantly fosters the election of female candidates, but only in the medium-run. This effect is driven by female candidates endorsed by the established political parties, and running for the first time. Finally, we uncover several mechanisms. Using survey data, we highlight a political demand mechanism: social media users, and especially the compliers of the Facebook access shock, have more favorable attitudes towards women and their presence in politics. Exploring the universe of Facebook posts of the candidates, we also document a political supply mechanism [work in progress]. Preliminary content analysis suggests gendered differences in online campaigning practices.

(with J. Loper and T. Taylor)

Can fertility policies have unintended effects on who gets married? We investigate the effect of the relaxation of the one-child policy, one of the strictest large-scale fertility policies of all time, on marriage. Before everyone was allowed to have two children with the two-child policy, some were already exempted from the strict one-child limit. Those exempted could have a "child advantage in the marriage market". In a context with a strong sex ratio imbalance with more men than women, the advantage can increase men's chances of getting married. Preliminary findings show that with the end of the one-child policy, those previously exempted as less likely to get married. The effect is exclusively driven by men -- the likelihood of women getting married stays unaffected. The results suggest that there was indeed a significant child advantage in the marriage market which, coupled with the skewed sex ratio, distorted who got married.

Co-author: Lucie Giorgi  

We explore the ability of conditional cash transfer programs to reduce child marriages in the presence of heterogeneous social norms. Contrary to popular belief, we show that groups strong child marriage norms can exhibit larger program effects for a given transfer than groups with weak norms. We empirically explore this relationship between norms and transfers by combining household data reflecting marriage ages with administrative data on one of the earliest and largest programs to reduce child marriage in North India. Consistent with our theoretical results, we find that regions and households with strong early marriage norms do exhibit large effects. Our results suggest that norms can be influenced by modest financial incentives depending on the configuration of norms and laws relative to household preferences. It is not therefore surprising that attempts to aim to change culturally driven behavior through cash transfers have had mixed success.

Evidence is lacking on how jobseeker search effort reacts to the expected level of discrimination they face in the labor market. Bridging this gap is important for understanding the channels through which discrimination affects final labor market outcomes. Using administrative data on job search and hiring from a large public job board, I show that an exogenous shock that plausibly increases the level of (perceived) discrimination towards minorities leads to a drop in minority search effort in the weeks that follow. This drop in search intensity is accompanied by an increase in match quality and increased hires in vacant jobs posted with the job board. I explore the role of selection and view the results through the lens of a directed search model in which jobseekers jointly determine the number of applications they emit and the types of jobs they apply to.

While economic factors in directed technical and scientific change have been widely studied, the role of teacher-directed scientific change has received less attention. This paper studies teacher-directed scientific change for one of the largest changes in the direction of research, the Scientific Revolution. Specifically, the paper considers the case of the English Scientific Revolution at the English universities of Oxford and Cambridge. It argues that exposure to different teachers shaped students' direction of research and can partly account for the successful trajectory of English science. For this, the paper introduces a novel dataset on the universe of all 111,242 students at English universities in the seventeenth and early eighteenth century and matches them to their publications. Using machine learning, the paper is able to quantify personal interest in different research topics. To derive causal estimates of teacher-student effects, the paper exploits a natural experiment based on the expulsion of fellows following the English Civil War and uses an instrumental variable design that predicts students’ choice of college based on their home regions. The paper finds strong empirical evidence of teacher-directed change in the English Scientific Revolution. These results illustrate how teacher-directed change can contribute to paradigm change.

How did Britain sustain faster rates of economic growth than comparable European countries, such as France, during the Industrial Revolution? We argue that Britain possessed an important but underappreciated innovation advantage: British inventors worked in technologies that were more central within the innovation network. We offer a new approach for measuring the innovation network using patent data from Britain and France in the 18th and early 19th century. We show that the network influenced innovation outcomes and then demonstrate that British inventors worked in more central technologies within the innovation network than inventors from France. Drawing on recently-developed theoretical tools, we quantify the implications for technology growth rates in Britain compared to France. Our results indicate that the shape of the innovation network, and the location of British inventors within it, help explain the more rapid technological change and industrial growth in Britain during the Industrial Revolution.

(joint with Walker Hanlon and Carl Hallmann)

We study the impact of conflict-driven displacement on human capital accumulation and occupational shifts. We focus on the Mozambican civil war (1977 − 1992), during which millions of civilians fled to the countryside, to cities, and to refugee camps in neighboring countries. We reconstruct the movements of the entire Mozambican population during this period to examine the consequences of multiple displacement trajectories in a unified setting. First, we describe the educational outcomes of the universe of individuals displaced during their school-going age. Second, we identify their age at move and implement a movers design to estimate exposure effects, focusing on moves occurring in high-conflict areas. We find that children who were still in their primary school-going years experienced strong and positive displacement effects on schooling. Third, we compare children of families separated during the war, using children who stayed behind as a counterfactual to each child’s displacement trajectory. Displacement is associated with increases in educational investments and shifts out of agricultural employment, with the largest effects experienced by rural-born children escaping to urban areas. The children who stayed behind do not report lower levels of schooling compared to their respective cohorts in their districts of birth. Fourth, we unpack the mechanisms and jointly estimate place-based and uprootedness effects. Fifth, we conduct a survey in Mozambique’s largest northern city, whose population doubled during the civil war: those displaced to the city have significantly higher education than their siblings in the countryside, and have converged to the levels of schooling of the non-mover urban-born individuals. However, those displaced exhibit significantly lower social/civic capital and have worse mental health, even three decades after the war ended. Our findings demonstrate that displacement shocks can stimulate human capital investments and break ties with subsistence agriculture. However, this comes at the cost of enduring social and psychological traumas.

We solve a real business cycle model with rational inattention (an RI-RBC model). In the RI-RBC model, the growth rates of employment, investment, and output are about as persistent as in the data, with an amount of inattention consistent with survey data on expectations. Moreover, consumption, employment, and output move in the same direction in response to news about future productivity. By contrast, the baseline RBC model produces neither persistent growth rates nor business cycle comovement after news shocks.

Relying on data from the US brewing industry, we implement two distinct approaches to estimate markups, using either production or demand data. We find that the two approaches produce similar estimates of average markups and both point to a significant upward trend. We combine the two approaches to evaluate the common (but controversial) assumption that retail markets operate competitively. Ultimately, our results suggest that demand-based markups recovered with the assumption of competitive retail markets are accurate, in our context, as long as downstream costs are properly accounted for.

(Joint with Paul T. Scott)

We study the history and geography of wealth accumulation in the US, using newly collected historical property tax records since the early 1800s. The US General Property Tax was a comprehensive tax on all types of property (real, personal, and financial), making it one of the first “wealth taxes.” Drawing on many historical records, we construct long-run, consistent, high-frequency wealth series at the county, state, and national levels. We first document the long-term evolution of household wealth in the US since the early 1800s. The US experienced extraordinary wealth accumulation after the Civil war and until the Great Depression. Second, we reveal that spatial inequality in the US has been large and highly persistent since the mid-1800s, driven mainly by Southern states, whose long-run divergence from the rest of the US predated the Civil War. Before the Civil war, enslaved people were assessed as personal property of the enslavers, representing almost one-half of total taxable property in Southern states. This system is morally abhorrent and implies wrongly counting forced labor income as capital. The regional distribution of wealth and the effects of the Civil war appear very different if enslaved people are not included in the property measure. Third, we investigate the determinants of long-term wealth growth and capital accumulation. Among others, we find that counties with a higher share of enslaved property before the Civil War or higher levels of wealth inequality experienced lower subsequent long-run growth in property.

In this paper we study preference heterogeneity by gender over job characteristics. Using information of both job postings and job seekers from an online job board, we document how job search behavior is strongly influenced by described job characteristics. We proceed in several steps: first, we define consideration sets of job ads for applicants using the network formed by job applicants, where each node is defined by joint applications by job seekers. Second, using text analysis of job descriptions, we identify jobs in terms of requirements and explicit diversity statements, along other characteristics. Third, we estimate application choice decisions by individuals on our entire sample. We document dimensions of jobs which attract females more than males, such as female-wanted, remote work, inclusive policies, and minorities recruiting. We also show that, while all individuals with higher wage expectations prefer inclusive policies, this is stronger for women. Finally, a social-interaction effects model also allows us to estimate that applicants prefer jobs with fewer expected applications, an effect that is larger for females. By providing evidence across the entire wage distribution, this paper differs markedly from most of the related literature, where either experimental settings or surveys of narrow segments of the labor market (e.g., economics undergraduates) may show only part of the story in terms of the root causes of gender differences in labor market outcomes.

This paper studies consumers' privacy choices when firms can use their data to make personalized offers. We first introduce a general framework of personalization and privacy choice, and then apply it to personalized recommendations, personalized prices, and personalized product design. We argue that due to firms' reaction in the product market, consumers who choose to share their data often impose a negative externality on other consumers. Due to this privacy-choice externality, too many consumers share their data relative to the consumer optimum; moreover, more competition, or improvements in data security, can reduce aggregate consumer surplus by encouraging more consumers to share their data.

Joint with Jidong Zhou.

We develop a spatial growth model for an agricultural economy where pollution diffuses in the soil. At each location, the only production factor is fertile soil, which is at the same time naturally bounded by the amount of available land, and eventually exposed to pollution diffused from neighboring locations. We develop a novel technique to obtain the policy maker’s optimal solution, which is analytical in the case of an homogeneous economy and covers all cases, for impatience rates ranging from almost zero to extremely high. When agents are very patient, the policy maker starts by making fully fertile all land before allowing for positive consumption. For slightly more impatient agents, the policy maker will allow for some consumption from the beginning, in the cleaning-up stage. With time, abatement stops, consumption raises and land becomes fully polluted in the long-term. We provide with some general results for the general spatially heterogeneous economy and its long-run, completing our study with some numerical exercises. Worth noting, simulations reveal that also in the non-homogeneous economy optimal consumption may transit through four different stages in time, responding to changes in fertile land and not necessarily in a smooth manner.

We study the interaction between monetary policy and labor supply decisions at the household level. We uncover evidence of heterogeneous responses and a strong income effect on labor supply in the left tail of the income distribution, following a monetary policy shock in the US and the UK. That is, while aggregate hours and labor earnings decline, employed individuals at the bottom of the income distribution increase their hours worked in response to an interest rate hike. Moreover, their response is stronger in magnitude relative to other income groups. We rationalize this using a two-agent New-Keynesian (TANK) model where our empirical findings can be replicated with a lower intertemporal elasticity of substitution for the Hand-to-Mouth households. This setup has important implications for the impact of inequality on the transmission of monetary policy. We unveil a novel dampening effect on aggregate demand generated by the Hand-to-Mouth substitution of leisure for consumption following a negative income shock. Therefore we show that the impact of inequality on the transmission mechanism of monetary policy is highly dependent on the different layers of heterogeneity on the household side and the different combinations of nominal and real frictions. More inequality does not necessarily generate a stronger response of aggregate demand after a monetary policy shock.

Young people in sub-Saharan Africa are particularly at high risk of HIV and other sexually transmitted infections. Using data from incentivized economic experiments conducted amongst 1,568 persons aged 15-29 years in Zimbabwe, we document the association between key individual preferences - risk aversion, altruism, present-bias and future-bias - at baseline and laboratory confirmed Herpes Simplex Virus type-2 (HSV-2) status and number of sexual partners, measured 12 months later. We find preferences are predictive of HIV risk behaviours. Women measured to be more altruistic at baseline were more likely to be HSV-2 positive 12 months later. Men measured to be risk averse at baseline were less likely to be HSV-2 positive at follow-up. There is limited association of preferences with number of sexual partners. We also show that preferences make a prominent relative contribution to predicting HIV risk behaviours, compared to other observable factors. Results highlight gender differences in the influence of preferences on HIV risk behaviours and potential for future research to design targeted interventions based on preferences.

In this paper we explore the effect of short-term memory on political outcomes in a model in which politics is viewed as a collective learning process. We analyse a dynamic model in which voters use past observations to make inferences about the true data generating process, and political parties are self-interested with polarised ideal policies. We model a probabilistic voting model where voters balance party loyalty with a desire to vote for the party whose policy provides higher expected utility. We show that short-term memory may lead to political cycles of polarisation and consensus. A short-term history involving only periods of consensus implies little variation in voters' data, and hence less precise knowledge about the true state of the world; this allows parties to push forwards with their self-interest. Alternatively, periods of polarisation imply sufficient variation which at some point allows voters to be confident about what is the right policy; this forces parties to converge on this policy. Our framework also sheds light on the relation between policy uncertainty and political polarisation, and on the effects of crises on political competition

We investigate the long-term economic consequences of mass refugee inflow. After the Greco-Turkish war of 1919–1922, 1.2 million Greek Orthodox were forcibly resettled from Turkey to Greece, increasing the host population by more than 20 percent within a few months. Building a novel geocoded dataset of more than 10,000 settlements in Greece, we examine the educational and occupational outcomes of refugee and native villages and their variation across birth cohorts throughout the 20th century. Using a Diff-in-Diff empirical approach, we find that refugee villages invested more in education and experienced greater structural transformation and industrialization relative to native ones. We find evidence in support for the “uprootedness” hypothesis, with forced migration (i) causing a shift in preferences toward investment in human portable assets (as opposed to physical) (ii) reducing the utility cost of leaving farming, as refugees are less attached to the land in their new location (relative to natives).

Joint with Nikos Benos, Stelios Karagiannis, Stelios Michalopoulos, Elias Papaioannou, and Seyhun Orcan Sakalli

We study the profitability of bundling by an upstream firm who licenses technologies to downstream competitors, and who faces competition for one of its technologies. In an otherwise standard “Chicago-style” model, we show that the existence of downstream competition can make inefficient bundling profitable. Forcing downstream firms to use a less efficient technology can soften competition, thus allowing the upstream firm to extract more profit through the licensing of its monopolized technology.

with Greg Taylor

Economic societies emerged during the late eighteenth-century. We argue that these institutions reduced the costs of accessing useful knowledge by adopting, producing, and diffusing new ideas. Combining location information for the universe of 3,300 members across active economic societies in Germany with those of patent holders and World’s Fair exhibitors, we show that regions with more members were more innovative in the late nineteenth-century. This long-lasting effect of societies arguably arose through agglomeration economies and localized knowledge spillovers. To support this claim, we provide evidence suggesting an immediate increase in manufacturing, an earlier establishment of vocational schools, and a higher density of highly skilled mechanical workers by mid-nineteenth century in regions with more members. We also show that regions with members from the same society had higher similarity in patenting, suggesting that social networks facilitated spatial knowledge diffusion and, to some extent, shaped the geography of innovation

In the literature on inequality and risk, changing distribution densities at the margin has been popularized as mean-preserving spreads by Rothschild and Stiglitz (1970). We import this idea to propose a definition of a positive productivity shock into the optimal nonlinear income tax framework à la Mirrlees. We then study the impact of a positive shock on productivity, taking place on a given interval I, on the optimal allocation and optimal marginal tax rates. We derive a general result for the class of rank-dependent social welfare functions consistent with Lorenz ordering (Simula and Trannoy, 2022a,b), including the maximin as limit case, with no condition on the elasticity of taxable income. Specifically, we show that the change in optimal marginal tax rates is non-monotonic on I. Our theoretical findings have meaningful empirical implications and can be used as a test of the optimal income tax model. We look at the fiscal situation in the US to provide an example where the predicted tax change according to the theory is verified, roughly speaking.

joint with Laurent Simula (ENS Lyon )

The Roma have faced poverty, discrimination and even enslavement and extermination since immigrating to Europe from India in the early 12thcentury. Persistent disparities between Roma and non-Roma throughout Europe have prompted recent integration initiatives. Drawing on a conceptual framework suggested by the literature on oppositional identity, aspirations, social capital, and the economics of religiosity, we explore the impacts of religious engagement on the integration of this marginalized population into broader society as captured by increased educational attainment. Using data from 12 Eastern European countries, we estimate the effect of religiosity on education with an instrumental variable (IV) approach that leverages exogenous variation in communist-era religious restrictions. We find that higher parental religiosity significantly increases child educational outcomes. Results indicate that religiosity also closes the gap between Roma respondents and their non-Roma neighbors for beliefs and other indicators. A placebo test among non-Roma households reveals no similar religiosity effects on educational outcomes of non-Roma children. Although more research is needed to identify specific mechanisms, these results suggest that religious engagement may soften oppositional identities in favor of alternative norms and aspirations. The role religious institutions may play in ultimately integrating the Roma and enhancing their future prospects merits further attention.

with Justin D. Kagin

Drawing on individual-level data from the 1911 and 1921 population censuses and census-linking techniques, we analyze the consequences of World War I in France on female labor force participation. Our results suggest that the war induced many women to enter the labor force: in municipalities that experienced greater military death rates, women were more likely to declare an occupation after the war than they did before the war. In contrast with results in the literature, in our sample, these effects are driven by women moving out of waged occupations outside the home (and inactivity) into farming activities.
joint with Lionel Kesztenbaum, PSE, INED)

We estimate behavioral responses to dividend taxation using recent French reforms: a rate hike and, five years later, a cut. Exploiting tax data at household and firm-level, we find very large dividend tax elasticities to both reforms. Individuals who control firms adjust dividend receipts instantaneously, accounting for most of the aggregate dividend reaction. Investment is insensitive to dividend taxation, except in small firms whose reaction is moderately negative. Dividend adjustments are instead driven by corporate saving, as owner-managers treat firms as tax-free saving vehicles. Small businesses’ profits decline following dividend tax increases, suggesting firms also serve as tax-free consumption vehicles.

Joint with Laurent Bach, Brice Fabre, Arthur Guillouzouic, Claire Leroy et Clément Malgouyres

Using multiproduct sales patterns and transitive production relationships across firms, we recover a continuous measure of distance between every pair of products and firms to uncover disjoint firm-product clusters. Within clusters, products closest to a firm are the most likely to be adopted and firms closer to potential new products grow and increase their scope faster. In higher sales clusters firms grow up with higher sales and less scope; in higher entrant clusters, firms grow out with lower sales and more scope. Firms with close competitors grow up more slowly but grow out more quickly, and are more likely to merge with these same competitors. Each extra rank of product distance decreases the frequency of adoption by one per cent relative to the base rate and one third of product adoptions are in the Top 10 closest products.

To what extent are media markets in democracies influenced by changes in political motives? We build a novel panel dataset of newspaper markets in India between 2002 and 2017 to measure the impact of changes in electoral incentives on how these markets develop over time. We exploit the announcement of an exogenous change in the boundaries of electoral constituencies to causally identify the relationship between the (future) electoral importance of news markets and the change in the number and circulation of newspapers. Using an event study and a staggered difference-in-difference approach, we show that markets that became more electorally important experienced a significant rise in both circulation and number of titles per capita. Both supply and demand seem to drive the increase, but we estimate that the former explains almost all the variation in the short run and around $60\%$ in the long run. Finally, we document how effects vary with prior levels of political competition and newspapers characteristics, and discuss implications for voting behavior and democratic accountability.

This document presents the outcome of two modules developed for the UK Foreign, Commonwealth Development Office (FCDO): 1) a forecast model which uses machine learning and text downloads to predict outbreaks and intensity of internal armed conflict. 2) A decision making module that embeds these forecasts into a model of preventing armed conflict damages. The outcome is a quantitative benchmark which should provide a testing ground for internal FCDO debates on both strategic levels (i.e. the process of deciding on country priorities) and operational levels (i.e. identifying critical periods by the country experts). Our method allows the FCDO to simulate policy interventions and changes in its strategic focus. We show, for example, that the FCDO should remain engaged in recently stabilized armed conflicts and re-think its development focus in countries with the highest risks. The total expected economic benefit of reinforced preventive efforts, as defined in this report, would bring monthly savings in expected costs of 26 billion USD with a monthly gain to the UK of 630 million USD.

Introducing a new measure of scientific proximity between private firms and public research groups and exploiting a multi-billion euro financing program of academic clusters in France, we provide causal evidence of spillovers from academic research to private firms. Private sector firms in the top quartile of exposure to the funding shock increase their R\&D effort by 20% compared to the bottom quartile. We then use qualitative evidence, exploiting reports produced by the funded clusters, as well as quantitative evidence, using administrative data on labor mobility and R&D public--private partnerships, to shed light on the channels for these spillovers. We show that spillovers are driven by contracting between the private and public sectors and, to a lesser extent, by labor mobility from one to the other and by informal contacts. We discuss policy implications of these findings.

(with Antonin Bergeaud, Arthur Guillouzouic and Emeric Henry)

University access has greatly expanded during the past decades and further growth figures prominently in political agendas. We study possible consequences of historical and future expansions in a stochastic, general equilibrium Roy model where tertiary educational attainment is determined by intelligence and disadvantage from low socioeconomic status or poor non-cognitive skills. The enlargement of university access enacted in the UK following the 1963 Robbins Report provides an ideal case study to draw lessons for the future. We find that this expansion led to the selection into college of progressively less intelligent students from advantaged backgrounds and to a declining college wage premium across cohorts. Our structural estimates indicate that the implemented policy was unfit to reach high-ability, disadvantaged individuals as Robbins had instead advocated. We show that counterfactual meritocratic selection policies would have attained that goal and so would have also been progressive.

(with Andrea Ichino and Aldo Rustichini)

We study, theoretically and empirically, how firms use inventories and alternative transport modes in the presence of demand and supply shocks. We build an sS model of an importer reselling an input subject to stochastic demand with fixed and variable input ordering costs that differ by the speed of shipment. We show that 1) U.S. supply chains outside of NAFTA can be managed like those in NAFTA using air shipments; 2) having the option to ship products by air is valuable and the value is proportional to the reduction in inventories; 3) large positive industry demand shocks are accommodated by a shift to air; 4) distant supply chains can better adjust to large industry demand shocks owing to the extra inventories on hand and the ability to shift to air shipment.

We study the effects and incidence of one of the world’s largest affirmative action programs in schooling: a 25% quota in all Indian private schools for disadvantaged students. Receiving a quota seat through lotteries increases private school attendance by 25 percentage points. Quota students also attend more expensive and preferred schools, on average. However, because applicants are disproportionately drawn from more-educated and better-off households within eligible castes, _67% of per-child expenditure is inframarginal for school choice. We investigate the causes of this regressive selection. Rich survey data on preferences show that low demand for private schools does not rationalize undersubscription. A follow-up randomized intervention that eased information constraints and application complexity increased application rates by 43% but did not reduce regressive selection (due to further unaddressed frictions). Our results highlight the distinction between undersubscription and regressive selection for policy design, and the importance of the joint incidence of multiple constraints in determining program take-up.
(joint with Mauricio Romero)

Poverty assistance is often administered in-kind even though cash transfers might raise recipients’ welfare more effectively. We characterize the political economy constraint that paternalistic motives impose on the welfare system. In our experiment, a representative sample of U.S. citizens reveal their motives by deciding whether to constrain real U.S. food stamp recipients’ choices between in-kind donations and cash equivalents we disburse. The modal respondent (40%) imposes the strictest possible constraints, while 30% impose no constraints. Hence the majority’s behavior is consistent with deontological motives rather than trade-off thinking. Yet, because of biased beliefs about recipient preferences, respondents underestimate the restrictiveness of their interventions, suggesting that they are partly misguided. When respondents decide between restricting recipients whom they believed should be restricted and giving choice to those whom they believe should have choice, they place substantially more weight on the former motive, which, too, tends to increase the severity of restrictions. Overall, respondents’ goal is not to ensure sufficient healthy nutrition, but to prevent consumption of items deemed inappropriate. While respondents reveal racial and gender stereotypes in various survey questions, neither donor nor recipient demographics have substantial effects on restriction decisions, though restrictions increase with respondents’ political conservatism. In-experiment behavior correlates strongly with views about government policy.

(joint with B. Douglas Bernheim, Tony Q. Fan, Zach Freitas-Groff)

Which firms and households will be most impacted by a carbon tax? To answer that question I set up a heterogenous agent, multi-sector model with putty-clay technology. A tax of $100 per ton of carbon emissions cuts emissions by 25% after 5 years, but it also reduces output by 1% in the short run and 1.75% in the long run. In the short term, the tax is slightly progressive despite poorer households spending more on carbon-intensive goods, the prices of which are rising. The putty-clay model’s inherent complementarity of capital and energy causes a sharp decline in capital income, a major source of income for top earners, and the resulting decline in investment causes job cuts in the capital goods-producing industries that employ high-income earners. In the longer run, as factor prices adjust, the tax becomes regressive and pre-tax real income in the lowest income percentile drops by 4%, 0.5 percentage points more than for the average household.

The equitable rate of interest represents a benchmark to evaluate the cross-sectional effects of monetary policy. I define it as the real rate of interest that minimizes the welfare losses associated to cross-sectional heterogeneity, under flexible prices. In a large class of models, it can be expressed as the payoff of a suitably chosen portfolio. In a Two-Agent New Keynesian model the deviations of the optimal policy prescription, relative to a Representative-Agent benchmark, can be traced back to the equitable rate gap: the difference between prevailing real rates and the equitable rate. This parallels the way in which the natural rate is the reference stick to evaluate the stance of monetary policy with regards to aggregate stabilization. Indeed, the difference between the natural rate and the equitable rate marks the tradeoff between aggregate and cross-sectional stabilization, faced by a welfare-maximizing policymaker.

Germany introduced the Gesellschaft mit beschränkter Haftung (GmbH) in 1892 in response to dissatisfaction with the menu of business enterprise forms then available to its entrepreneurs. The GmbH combines some features of the corporation with the flexibility of the partnership, and can be compared to vehicles such as the Limited Liability Company (LLC) popular in the United States today. The new form quickly became popular and is now the most common enterprise form in Germany. In creating this new legal form, the legislator engaged in a rare form of institutional innovation; other enterprise forms such as the partnership and corporation date to at least medieval times. Pre-introduction arguments reflected assumptions about how business firms would use the GmbH. This paper compares the predictions to the reality for the period prior to 1914. GmbHs became more popular than most expected. Critics of the new form exaggerated some practices they viewed as abuses, probably because they relied on specific examples of problem firms rather than comprehensive data. The lower contracting costs implicit in the GmbH led to its wide use in many sectors and sometimes for entirely new purposes.

Flexible repayment benefits borrowers, but practitioners fear diminished repayment morale. We study repayment choices in rigid and flexible loan contracts with discretion in repayment timing. Using a lab-in-the-field experiment with microcredit borrowers in the Philippines, we identify moral hazard and quantify social pressure. In our rigid benchmark contract, repayment is higher than payoff maximization predicts. Flexibility substantially lowers both repayment and social pressure. Our results are consistent with a strong social norm for repayment, which is weakened by introducing flexibility. This suggests that cooperative behavior determined by social norms may erode if the applicability of these norms is not straightforward.

With real interest rates below the growth rate of the economy, but the marginal product of capital above it, the public debt can be lower than the present value of primary surpluses because of a bubble premia on the debt. The government can run a deficit forever. In a model that endogenizes the bubble premium as arising from the safety and liquidity of public debt, more government spending requires a larger bubble premium, but because people want to hold less debt, there is an upper limit on spending. Inflation reduces the fiscal space, financial repression increases it, and redistribution of wealth or income taxation have an unconventional effect on fiscal capacity through the bubble premium.

We establish a set of novel empirical facts concerning cross-section distributions of inflation expectations reported in surveys. Almost all the variation in expectations about their mean may be summarized via three factors we call disagreement, skew, and shape. We adopt a functional principal component regression approach to estimating forward-looking models of inflation that exploits the heterogeneity present in individual-level data. By using survey information more effectively, our approach reveals an enhanced role for expectations in inflation dynamics that is robust to lagged inflation, trend inflation, and supply factors. Our findings hold in similar form across two major economies.

Since the end of 2008, the Federal Reserve has b een communicating its monetary p olicy in terms of two instruments under its direct control: the interest rate on bank reserves (IOR rate), and the size of its balance sheet. We intro duce banks and bank reserves into the basic New Keynesian mo del to assess the main consequences of this p olicy change. We show that our mo del can account, in qualitative terms, for three key features of US inflation during the 2008-2015 zero-lower-bound (ZLB) episo de: no significant deflation, little inflation volatility, and no significant inflation following quantitative-easing p olicies. Crucial to this result is our assumption that demand for bank reserves got close to satiation, but did not reach full satiation. We intro duce liquid government b onds into the mo del to reconcile our non-satiation assumption with the fact that Treasury-bill rates dropp ed b elow the IOR rate during the ZLB episo de. Lo oking ahead, we explore the implications of our mo del for the normalization of monetary p olicy and its future op erational framework (floor vs. corridor system). In particular, we find that current and expected future IOR-rate hikes and balance-sheet contractions are always deflationary in our model, thus ruling out Neo-Fisherian effects

joint with Behzad Dibat

This paper explores the link between infrastructures built by autocratic regimes and political values in the wake of the transition to democracy and in the long run. In Fascist Italy (1922-43), Mussolini founded 147 “New Towns” (Città di Fondazione). Exploring municipality-level data before and after their construction, I document (i) that the New Towns enhanced local electoral support for the Fascist Party and (ii) that the effect persisted through democratization, enhancing local support for Italy’s neo-fascist party, which endured until recent times. Placebo estimates of New Towns planned but not built and spatial regression discontinuity design both support a causal interpretation of this pattern. Survey respondents near the New Towns currently exhibit preferences for a stronger leader in politics, for nationalism, and for the fascists as such. The effect is greater for individuals who lived under the Fascist Regime and is transmitted across generations inside the family. The findings suggest that authoritarian leaders may exploit public investment programs to induce a favorable view of their ideology, which persists across institutional transitions and over the long term.

Can firms exploit behavioral biases to increase profits? Does consumer sophistication about these biases limit the scope of exploitation? To answer these questions, I run a series of natural field experiments with over 600,000 consumers and estimate novel sufficient statistics of consumer sophistication. The empirical application is a ubiquitous and widely regulated form of price discrimination: rebates that need to be actively claimed by consumers. These promotions are suspected of boosting sales even though many consumers eventually fail to claim the rebate—a phenomenon marketers refer to as “slippage.” I show theoretically that consumers’ subjective redemption probabilities can be inferred from how demand responds to rebates as opposed to simple price reductions. I identify these elasticities in three natural field experiments with a major online retailer, in which I randomize prices, redemption requirements, and reminders. Results reveal that claimable rebates in fact increase sales substantially even though 47% of consumers do not redeem the rebate. However, consumers exhibit a remarkable degree of sophistication: the demand response to a rebate is only 76% of the demand response to an equivalent price reduction. Structural estimates imply that consumers are almost perfectly aware of their inattention but vastly underestimate the hassle of redemption by 20 EUR per consumer. Exploiting this misperception increases the profitability of rebates by up to 260%.

The paper can be found here: https://drive.google.com/file/d/1zgH-BDx-psjSHzDGVIx4kfNB2HBg1EMe/view

We investigate impacts of antidepressant use in childhood on academic achievement at age 16 using linked register data on individuals in Denmark. Leveraging quasi-random assignment of patients to child psychiatrists with different prescribing tendencies, we find that antidepressant treatment significantly increases test scores. Treatment effects are larger and only precisely estimated for math (rather than Danish) and for girls (rather than boys). Marginal treatment effects are positive across a large part of the distribution, and the slope of the MTE curve reveals selection on unobservables into gains. Investigating heterogeneity in treatment effects on observables, we find the reverse: treatment effects are driven by the children of less educated mothers but, taking all cases rather than the marginal cases picked out by the instrument, these families are less likely to be treated. Overall, our results indicate large detrimental impacts of mental health disorders on cognitive performance, and high returns to antidepressant treatment. They are consistent with evidence that depression and anxiety have a larger impact on math performance, and that girls are more sensitive to these conditions.

Joint with Meltem Daysal, Mircea Trandafir, Nis Vestergård Lydiksen

We train a random sample of small informal Ghanaian cassava processors on two simple practices: production measurement and goal setting. The former is essential for any firm to survive, let alone grow. The latter is an ubiquitous practice in the western world and a proven method to enhance productivity. Despite their importance, these practices are not systematically used by small informal firms in the developing world. While neither practice requires substantial material resources to be implemented, both may require a change in mind-set and deep-rooted norms. We follow the cassava processors for two months after training, and collect daily data on production and goals. We find a significant positive effect of goal-setting on productivity. Firms trained in goal-setting increase their productivity by 50% relative to those trained in production measurement only. In particular, male employers and employers who are less well-educated, with less experience in goal-setting, with smaller firms, and who are more impatient and more risk-averse set higher goals. Finally, we observe that, on average, workers systematically underachieve their daily production targets to a moderate extent, which suggests that goals are kept rather high as a motivational device. All in all, we confirm that goal setting can be an elective and inexpensive tool to increase productivity amongst small informal enterprises in non-western cultures.

The prevalence of metabolic risk factors that contribute to cardiovascular disease (CVD), such as hypertension and high plasma glucose levels have seen a substantial increase globally. Violence elevates stress and increases CVD risk and yet, there is sparse evidence on the relationship between domestic violence and CVD risk factors. This study presents new empirical evidence by leveraging biomarker data from a large nationally representative survey. I find (i) a consistent positive causal effect of physical violence on prevalence of hypertension amongst women, (ii) a positive causal effect of emotional and sexual violence on prevalence of hypertension amongst women, (iii) No discernible effects of domestic violence on CVD risk for the men in these relationships.

We propose a cross-nested logit (CNL) approach to address the question of how people revise their migration decisions when facing changes in the global environment. Compared with the popular logit model, the CNL allows for more sophisticated substitution patterns between destinations. Using migration aspiration data from India, we show that the CNL approach outperforms competing approaches in terms of quality of fit and predictive power, implies stronger heterogeneity in responses to shocks, and highlights complex and intuitive substitution patterns. We document the low degree of substitutability between the home and foreign alternatives as well as within subgroups of particular countries. This approach allows to capture spillover effects of a given immigration policy on immigration pressures faced by other countries of destination.

Joint with Michel Bierlaire and Frédéric Docquier.

Research is increasingly a collaborative team exercise involving multiple researchers, yet little is known about how the composition of such teams affects their research output. This paper examines how the gender composition of research teams influences their performance. We take advantage of a field experiment in which first-year economics students are randomly paired together and perform research-like tasks. We find large differences in research performance, as measured by the grades they receive, by gender composition. All-male teams are significantly outperformed by both mixed and all-female research teams. These differences remain even when comprehensively controlling for the individual research aptitude of the group members. No comparable compositional effect is found for other characteristics, such as ethnicity and socio-economic status.

Joint with Max Coveney, Teresa Bago d'Uva

How can the government move the economy from a secular stagnation equilibrium, with under-employment and a permanently binding zero lower bound, to a neoclassical equilibrium, with full employment? This can be achieved through a temporary, but massive, fiscal stimulus to overheat the economy such as to raise the inflation anchor. Despite the substantial cost of overheating the economy, this policy is typically optimal. The lack of fiscal space cannot prevent the government from pump priming the economy through fiscal policy. It may in fact help spur inflation. To keep a tight control over the price level, government debt should have a sufficiently long maturity.

Should cash transfer programmes restrict consumer choice? For example, should food assistance delivered in cash be restricted to food and exclude temptation goods? Theoretically, Southworth (1945) showed that restrictions induce (1) a substitution effect away from restricted goods and (2) a negative wealth effect if transfers are extra-marginal and resale of goods is costly. The welfare impact on transfer recipients is negative. We test Southworth’s theory by exploiting a natural experiment in a refugee settlement in Kenya, where some refugees receive monthly cash transfers restricted to food while others get unrestricted cash transfers. We corroborate Southworth’s predictions empirically: restricted transfers increase participation in a shadow resale market and negatively affect non-food expenditure, temptation-goods spending, and subjective well-being. Consistent with theory, restrictions have no significant effect on food consumption. Our results show that policy-makers should avoid restrictions to maximise positive impacts on transfer beneficiaries.
With Jade Siu and Cory Rodgers

In this paper we explore the effect of short-term memory on political outcomes in a model in which politics is viewed as a collective learning process. We analyse a dynamic model in which voters use past observations to make inferences about the true data generating process, and political parties are self-interested with polarised ideal policies. Voters balance party loyalty with a desire to vote for the party whose policy is a based on a better interpretation of past observations. We show that short-term memory leads to political cycles of polarisation and consensus. A short-term history involving only periods of consensus implies little variation in voters’ data, and hence less precise knowledge about the true state of the world; this allows parties to push their self interests. Alternatively, periods of polarisation imply sufficient variation which at some point allows voters to be confident about what is the true model; this forces parties to converge on the policy that fits that model. Our framework also sheds light on the relation between policy uncertainty and political polarisation, and on the effects of crises on political competition.

"Paid time off is one of the most common and most valued amenities of employment contracts. This study analyses the labor market effects of mandating paid time off. We leverage a German court ruling that mandated more time off for specific age groups in some industries and find that the ruling significantly increased paid time off, with no adverse effects on employment or wages. We interpret these results through the lens of canonical labor market models and find evidence for imperfect competition in the labor market. Our results suggest a similar degree of market power of employers over low and high-paid workers."

This paper investigates the importance of the age composition in the Covid-19 pandemic. We augment a standard SIR epidemiological model with individual choices on work and non-work social distancing. Infected individuals are initially uncertain unless they are tested. We find that older individuals socially distance themselves substantially in equilibrium. Con-fining the old even more reduces their welfare. Confining the young ex-tends the duration of the epidemic, with negative consequences on the old if the epidemic cannot be controlled after confinement. Testing and quarantines save lives, even if conducted just on the young, as does separation of activities by age. Combining policies can increase the welfare of both the young and the old

We develop a search-and-matching model where the magnitude of unemployment insurance benefits affects the likelihood that unemployed actually engage in active job search. To quan- titively discipline this relation we use administrative data of unemployed search audits. We use the model to quantify the effects of unemployment reforms. For small benefits' increases, the policymaker faces a trade-off between an uptick in the measure of unemployed actually searching and a fall in the unemployment exit-rate conditional on searching. For larger bene- fits' increases, an active search margin magnifies the benefits' disincentives, leading to a bigger drop in the employment rate than previously thought.

Interviewing is a decisive stage of most processes that match candidates to firms or organizations. This paper studies how the interview assessment of a candidate depends on the other candidates seen by the same evaluator, and their relative timing in particular. We leverage novel administrative data covering about 29,000 one-to-one interviews conducted within the admission process of a prestigious study grant program in Germany. Identification relies on the quasi-random assignment of candidates to evaluators and time slots. We find that a candidate's assessment decreases when her evaluator receives a better candidate draw. Moreover, the influence of the previous candidate is about three times stronger than the influence of the average other candidate in the sequence. The empirical pattern suggests that evaluators exhibit a contrast effect caused by the interplay between the associative recall of prior candidates and the attention to salient quality differences.

Interviewing is a decisive stage of most processes that match candidates to firms or organizations. This paper studies how the interview assessment of a candidate depends on the other candidates seen by the same evaluator, and their relative timing in particular. We leverage novel administrative data covering about 29,000 one-to-one interviews conducted within the admission process of a prestigious study grant program in Germany. Identification relies on the quasi-random assignment of candidates to evaluators and time slots. We find that a candidate's assessment decreases when her evaluator receives a better candidate draw. Moreover, the influence of the previous candidate is about three times stronger than the influence of the average other candidate in the sequence. The empirical pattern suggests that evaluators exhibit a contrast effect caused by the interplay between the associative recall of prior candidates and the attention to salient quality differences.

We propose a new method of eliciting individual time preference measures in settings where income or consumption changes over time. Standard time preference measures elicit the marginal value of money across different periods of time. If income remains constant, in standard models this identifies the discount factor. If income is volatile (and perfect smoothing infeasible), it measures the discount factor and changes in the marginal consumption utility. Our method relies on allocating lottery tickets with low winning probabilities but high rewards. In standard intertemporal choice models, the high reward decouples the allocation of lottery tickets from the current and expected future income levels (i.e., from changes in wealth and other income over time). We validate our method on experimentally on two student samples drawn from the same pool. One set of students is asked about discount factors in December at a time when their current budget is reduced by extraordinary expenditures for Christmas and Saint Nicholas gifts. The other one is asked in February when no such extra constraints exist. Finally, we apply our method to elicit discount factors from unemployed job seekers which naturally have varying income streams.

We provide a canonical and tractable model of a trade platform which enables buyers and sellers to transact and charges commission for hosting third party products. We extend it for “hybrid" platforms (like Amazon), which also resell their own products. The hybrid platform controls both the number of differentiated products (variety) it hosts and their prices via its own product price and commission. We micro-found the model by engaging three key recent conceptual innovations: a mixed oligopoly demand model where a large player (the platform’s product) interacts with a monopolistically competitive fringe (3rd party sellers); long-run aggregative games with free entry to model seller entry; Weitzman's optimal search rule adapted to where buyers choose to participate on-board the platform depends on both heterogeneous outside options and joining costs. Allowing the platform sell its own product (hybrid mode) hurts consumers. The larger the market share of the platform's product, the higher the platform's commission on 3rd-party sales and the lower are consumer benefits. We support our theoretical predictions with some evidence from Amazon.

We identify a new mechanism through which cultural diversity affects economic outcomes, based on a model of culture as shared cognition. Under this view, cultural diversity matters because it increases strategic uncertainty. The model can help better understand a variety of disparate evidence, including why homogeneous societies can be more conformist, why diverse societies may get stuck in a low-trust trap, why companies with a strong culture may fail to adopt superior work practices, and why autocratic rulers in diverse societies may overinvest in state capacity.

We derive direct inference on the macroeconomic effects of helicopter money drops using the random allocation of the Spanish Christmas Lottery in different provinces, as a natural experiment. This is the only case in the world where individuals living in one region randomly receive pure monetary transfers of a considerable amount. In particular, we study the monthly regional level data on unemployment from the National Employment Agency (SEPE) and CPI prices from the Spanish Statistical Office (INE) and individual and regional data on consumer sentiment regarding the Spanish and individual economic outlook drawn from the Spanish Center of Sociological Research (CIS). We show, using local projections, that the money drops reduce unemployment and increase CPI prices at the regional level. Money drops significantly boost economic sentiment for both current and future economic conditions both at the province and at the individual level and their effects are stronger during recessions. Such policies impact mostly the sentiment of younger, less educated and poorer individuals and women.

This paper evaluates the labor market effects of sick pay mandates in the United States. Using the National Compensation Survey and difference-in-differences models, we estimate their impact on coverage rates, sick leave use, labor costs, and non-mandated fringe benefits. Sick pay mandates increase coverage significantly by 13 percentage points from a baseline level of 66%. Newly covered employees take two additional sick days per year. We find little evidence that mandating sick pay crowds-out other non-mandated fringe benefits. We then develop a model of optimal sick pay provision along with a welfare analysis. For a range of plausible parameter values, mandating sick pay increases welfare.

Joint with Johanna Catherine Maclean and Nicolas R. Ziebarth

New technologies make it easier for charismatic individuals to influence others. This paper studies the political impact of the first populist radio personality in American history. Father Charles Coughlin blended populist demagoguery, anti-Semitism, and fascist sympathies to create a hugely popular radio program that attracted tens of millions of listeners throughout the 1930s. Exploiting variation in the radio signal strength as a result of topographic factors, I find that a one standard deviation increase in exposure to Coughlin’s anti-Roosevelt broadcast reduced FDR’s vote share by about two percentage points in the 1936 presidential election. Effects were larger in counties with more Catholics. An alternative difference-in-differences strategy exploiting Coughlin’s switch in attitude towards FDR during 1932-1936 confirms the results. Moreover, I find that places more exposed to Coughlin’s broadcast in the late 1930s were more likely to form a local branch of the pro-Nazi German-American Bund and sold fewer war bonds during WWII.

This paper investigates whether political connections affect individuals’ propensity to engage in white-collar crime. We identify connections by campaign donations or direct friendships and use the 2007 French Presidential election as a marker of change in the value of political connections to the winning candidate. We compare the behavior of Directors of publicly listed companies who were connected to the future President to the behavior of other non-connected Directors, before and after the election. Consistent with the belief that connections to a powerful politician can protect someone from prosecution or punishment, we uncover indirect evidence that connected Directors are more likely to engage in suspicious insider trading after the election: Purchases by connected Directors trigger larger abnormal returns, connected Directors are less likely to comply with trading disclosure requirements in a timely fashion, and connected Directors trade closer in time to their firms’ announcements of results.

Joint with Thomas Bourveau and Renaud Coulomb

Monitoring economic conditions in real time,or nowcasting, is among the key tasks routinelyperformed by economists. Nowcasting entails some key challenges, which also characterisemodern Big Data analytics, often referred to as the three “Vs”: the large number of timeseries continuously released (Volume), the complexity of the data covering various sectors ofthe economy, published in an asynchronous way and with different frequencies and precision(Variety), and the need to incorporate new information within minutes of their release (Ve-locity). In this paper, we explore alternative routes to bring Bayesian Vector Autoregressive(BVAR) models up to these challenges. We find that BVARs are able to effectively handlethe three Vs and produce, in real time, accurate probabilistic predictions of US economicactivity and, in addition, a meaningful narrative by means of scenario analysis.

Joint with Jacopo Cimadomo, Domenico Giannone, Michele Lenza and Andrej Sokol

What does preference heterogeneity imply for consumption inequality? This paperstudies the link from wage to consumption inequality within a lifecycle model of con-sumption and family labor supply. Its distinctive feature is that households have generalheterogeneous preferences over consumption and labor supply. The paper shows iden-tification of the joint distribution of unobserved household preferences separately fromthe observed distributions of incomes and outcomes. Estimation on data from the PanelStudy of Income Dynamics in the US reveals substantial unexplained heterogeneity inconsumption preferences but little unexplained heterogeneity in labor supply preferences.Consumption preference heterogeneity accounts for about a third of consumption inequal-ity in recent years and implies a distribution of partial insurance of wage shocks.

I measure the uncertainty affecting estimates of economic inequality in theUS and investigate how accounting for properly estimated standard errors can affect theresults of empirical and structural macroeconomic studies. In my analysis, I rely upon twodata sets: the 1989–2019 Survey of Consumer Finances (SCF), which is a triennial surveyof household financial condition, and the 1991–2012 Individual Tax Model Public Use File(PUF), an annual sample of individual income tax returns. While focusing on six income andwealth shares within the top 10 percent, my results suggest that ignoring uncertainties inestimated income and wealth shares can lead to statistically imprecise conclusions about thepast and current levels of top income and wealth inequality, and therefore, lead to inaccuratepredictions and potentially ineffective policy recommendations. My analysis suggests thatfor the six top-decile income shares under consideration, the PUF estimates are considerablybetter than those constructed using the SCF; for wealth shares of the top 10 to the top 0.5percent, the SCF estimates appear to be more reliable than the PUF estimates; finally, forthe two most granular wealth shares, the top 0.1 and 0.01 percent, both data sets presentnon-trivial challenges that cannot be readily addressed.

I develop a hypothesis test for model evaluation which is robust to time-variation in parameters. The proposed method can be applied in-sample and out-of-sample to any economic model based on moment conditions. In-sample, the test selects between two nested model specifications in the presence of parameter instabilities. Out-of-sample, the test can be used to evaluate the performance of model or judgmental forecasts robust to time-variation. The key feature of the proposed test is that it is particularly powerful in the presence of multiple shifts in parameters without imposing a specific form of time-variation. Further, the test statistic provides narrative evidence on which parts of the sample drive the rejection of the null hypothesis. Simulations show that the test is accurately sized in finite samples and is more powerful than tests assuming constant coefficients or a single break if the data-generating process exhibits multiple shifts in parameters. Using the proposed test, I document the presence of short-horizon predictability in the U.S. equity premium during the postwar period. I find evidence of predictability for a large set of variables once timevariation is taken into account. The test further provides evidence of heterogeneity in the location of predictability episodes across variables. The findings explain why traditional tests often fail to uncover predictability in the full sample and why studies that split the sample at different dates often arrive at conflicting results regarding the predictive ability of a wide class of variables.

The objective of this paper is to calculate asylum seekers’ex antereturns onoverstaying when their application is rejected. It takes advantage of a uniquesurvey designed to permit estimation of a structural model about the decision tooverstay. The proposed model sees this decision as a utility maximization problemwith uncertainty about the final choice. The paper proposes mild conditions fornonparametric identification of several objects of interest, including populationdistribution ofex antepecuniary benefits, non-pecuniary costs, surpluses, andoption-value created by the chance of becoming regularized in the future. Estimationis conducted using a semiparametric estimation procedure on a sample of Afghanasylum seekers.Ex antesurpluses of overstaying are predominantly positive butvery heterogeneous in the population. The pecuniary benefits explain a modestpart of these surpluses. In contrast, the option-value and the cost associated witha deportation are important components of the expected returns. Afghan asylumseekers are ready to spend a long time with a precarious status to eventually obtainthe right to stay and the amenities associated with it.

This paper considers financial markets with bid-ask spreads and studies the class of markets with hedging complements, a property formalized by the complementarity of its hedging price, in the same way as strategic complements is defined on agents’ payoff functions in game theory. The class of markets with hedging complements contains both markets with frictionless securities and the larger class markets with independent marketed securities together with the frictionless bond, assuming both to be arbitrage-free. Moreover, the hedging prices of the latter markets are proved to satisfy a tractable explicit formula, as the sum of a “generalized” convex Choquet integral and of a modular term. Finally this class of markets also satisfyg the put-call parity of Cerreia-Vioglio, Maccheroni, Marinacci, and Montrucchio (2015).

Recently a credibility crisis has taken hold across the social sciences, arguing that a component of Fischer (1935)’s tripod has not been fully embraced: replication. The importance of replications is not debatable scientifically, but researchers and journals’ incentives are not sufficient to encourage replications. We analyze a novel, decentral- ized approach promoting replications through beneficial gains between scholars and editors. We highlight the trade-offs involved in seeking independent replications before submission to journals, and demonstrate the operation of this method via an investi- gation of the effects of Knightian uncertainty on cooperation in public goods games, a pervasive but largely unexplored feature in the literature.

Aspirations for a better future can motivate people to make long-term investments. But do aspirations constrain investment only when people are poor, or do they continue to limit investment when people become wealthier? We run a 415-village field experiment in Western Kenya to study this relationship. We cross-randomize two interventions: a video documentary about successful people in similar villages who escaped poverty, which aims to increase aspirations; and a 2,273 USD PPP unconditional cash transfer. We find that aspirations limit investments: the psychological intervention alone increases aspirations, labour supply, expenditure on inputs, revenue, and education spending, by roughly one quarter to one third as much as the cash transfer. After people receive cash, their aspirations for income and wealth increase, and aspirations no longer constrain investments in economic activities. However, cash alone does not increase aspirations for and investments in children's education. In contrast, the group who receive both video and cash have higher education aspirations and invest more in children's education than the group who receive only cash. These results suggest that increases in wealth alone remove most of the constraints of aspirations on investment, but low aspirations may still limit longer-term, potentially high-return investments

A large literature on social networks shows that individuals are more likely connected with each other if they have common connections with other individuals. We take this idea to trade between firms. Firms are more likely to trade with each other if they share common trading partners, and forming a new connection makes trade with other partners more likely. We embed such transitivity effects in a model of trade in differentiated products. Transitivity effects act as an externality leading to lower effective trade costs, especially over short distances where trade connections tend to be dense. This implies an agglomeration-like impact at the local level. We show that transitivity effects between firms translate to the aggregate level of trade between locations. In general equilibrium, the model yields a gravity framework with endogenous bilateral trade costs that depend on the intensity of shared trade links with third locations. On the empirical side, we use data on domestic US manufacturing shipments to estimate transitivity effects. Our results suggest that they can account for a sizeable fraction of the variation in trade flows.

We analyse the existence and underlying mechanisms of neighbourhood effects in welfare-to-work transitions in the context of a dominant proportion of foreign-born individuals. The analysis is based on Luxembourg social security longitudinal data, covering 2001–2015, and provides precise information at the postcode level, corresponding mostly to streets. Our identification strategy exploits the exogenous variations provided by the very fine data granularity and follows two paths. We first examine interactions among all neighbours using an individual-level analysis, before focusing on interactions among only welfare recipients using a matched-pair analysis. This second step allows us to deal with the mediating effect of welfare recipients’ citizenship. The main findings highlight the existence of neighbourhood effects in welfare-to-work transitions, which are also affected by the characteristics of the neighbours, including their citizenship. These characteristics suggest that social norms and/or stigma, prevail over the support for welfare recipients to find a job, and over the in-group support for welfare recipients. The matched-pair analysis provides contrasting results across citizenship for individuals from large-sized citizenship groups (interactions within the own group) and individuals from medium-sized groups (interactions between groups).

Joint with Alessio Fusco

We study optimal time-consistent monetary and fiscal policy in a New Keynesian model where occasional declines in agents' confidence give rise to persistent liquidity trap episodes. Insights from widely-studied fundamental-driven liquidity traps are not a useful guide for enhancing welfare in this model. Raising the inflation target, appointing an inflation-conservative central banker, or allowing for the use of government spending as an additional stabilization tool can exacerbate deflationary pressures and demand deficiencies during the liquidity trap episodes. However, appointing a policymaker who is sufficiently less concerned with government spending stabilization than society eliminates expectations-driven liquidity traps.

We consider a model in which firms sell differentiated products, and consumers are interested in buying repeatedly over time but need to search for price and product information. Firms and consumers turn over at an exogenous rate. We show that provided the search cost is not too large, the market exhibits pure strategy price dispersion. Specifically, older firms charge higher prices because they face a larger and `better-matched' demand. The fact that sellers gradually raise their prices over time also leads to rich consumer search and purchase dynamics. For example, consumers may initially search a lot for a product, return to the seller and buy for several periods, but then faced with successive price increases quit the firm and search again for a new product. We also provide conditions under which the ability of sellers to contact past customers and offer them personalized prices leads to higher consumer surplus.

We examine the effect of a physician network on medical innovation using novel matched patient-physician-hospital panel data. The data include every relevant physician and all patients in the English NHS for 15 years and physicians’ workplace histories for more than 20. The dynamic network arising from physician mobility between hospitals over time allows us to separate unobserved physician and hospital heterogeneity from the effect of the network. We build on standard peer-effects models by adding cumulative peer behaviour and allow for particularly influential physicians (‘key players’), whose identities we estimate. We find positive effects of peer innovation take-up, number of peers, and proximity in the network to both pioneers of the innovation and key players. Counterfactual estimates suggest that early intervention targeting young, connected physicians with early take-up can significantly increase aggregate take-up.

We provide new evidence on how conflict adversely affects economic outcomes. Specifically, we ask whether and how the production network is a first-order determinant of the propagation of conflict to firms outside of conflict zone. Using microdata on Indian manufacturing plants and geo-coded information on Maoist insurgency, we first provide an estimate of the direct exposure to conflict. Firms located in conflict affected areas suffer a loss of 7-11% of their output. Estimating structurally a general equilibrium model of production networks, we then obtain an estimate for the overall macroeconomic impact of the Maoist insurgency by taking this propagation effects into account. We find that the Maoist insurgency resulted in a 0.4-0.7% decline in aggregate output of Indian’s manufacturing sector. Only the 20% of this loss is due to direct exposure to conflict, whereas the remaining 80% explained by the indirect exposure to conflict through the network production.

 

Motivated by the growing discussion on the legitimacy of multilevel marketing schemes and their resemblance to pyramid scams, we compare the two phenomena based on their underlying compensation structures. We identify the conditions on the tendency of agents to spread information by word of mouth, under which a principal can design a pyramid scam to exploit a network of agents whose beliefs are coarse. We find that a pyramid scam is sustainable only if its underlying reward scheme compensates each participant based on at least two levels of recruitment, i.e., the people he recruits and the people they recruit.

It is often argued that forming a monetary union increases financial integration and leads to a more efficient allocation of capital across member countries. To investigate this claim, this paper provides a model connecting monetary policy and international capital flows. The key feature of the model is that  monetary policy affects the value of collateral that can be repossessed by creditors in case of default.  Once a country has accumulated a large stock of  external debt,  it is optimal for the government  to depreciate the exchange rate to expropriate foreign investors.  Anticipating this, under flexible exchange rates international investors  impose tight borrowing limits on foreign debtors.  In a monetary union this source of exchange rate risk is absent, because national governments do not control monetary policy. Forming a monetary union thus increases financial integration by boosting borrowing capacity toward foreign investors. This process, however, does not necessarily lead to higher welfare. Higher financial integration may increase welfare by fostering capital flows from capital-abundant to capital-scarce countries. But a high degree of financial integration may also make episodes of inefficient capital flights out of capital-scarce countries possible.

 

Portfolio under-diversification is one of the most pervasive phenomena in household finance. We show that salience theory of choice under risk synthesizes seemingly disparate explanations for under diversification that have been established in the literature: skewness preferences and correlation misperception. On the one hand, we study under-diversification with symmetric assets and find that salience theory can rationalize experimental evidence on naive diversification. On the other hand, we derive predictions on the role of correlation for portfolio selection with skewed assets, which allow us to distinguish between our salience-based explanation of under-diversification and naive decision rules such as the 1/N-heuristic. We plan to test our novel predictions in a lab experiment.
     

We study incumbency advantage in platform industries, where the utility of participating in a platform is increasing in the mass of users participating in that platform. Individuals receive stochastic opportunities to migrate from an incumbent to a new (entrant) platform, which they can accept or wait until the next opportunity arises. Individuals have an incentive to delay migration until enough other users have migrated, which provides a micro-foundation for incumbency advantage. When users obtain more frequent migration opportunities, the cost of delaying migration is reduced, so incumbency advantage increases. Migration technologies that allow for large groups of individuals to migrate in a short period of time (i.e., coordination) are also associated with higher incumbency advantage. There always exists some capacity constraint by the entrant which increases the cost of delaying migration and thereby reduces incumbency advantage. Multi-homing reduces incumbency advantage but does not eliminate it. When individuals have heterogeneous preferences for the two platforms, there can be welfare losses due to excessive segregation of individuals across the platforms.

 

The paper re-examines whether the Federal Reserve's monetary policy was a source of instability during the Great Inflation by estimating a sticky-price model with positive trend inflation, commodity price shocks and sluggish real wages. Our estimation provides empirical evidence for substantial wage-rigidity and finds that the Federal Reserve responded aggressively to inflation but negligibly to the output gap. In the presence of non-trivial real imperfections and well-identified commodity price-shocks, U.S. data prefers a determinate version of the New Keynesian model: monetary policy-induced indeterminacy and sunspots were not causes of macroeconomic instability during the pre-Volcker era.

Do political leaders alter the course of history by changing political preferences and behaviours? We answer this question by showing how World War I combat exposure to Philippe Pétain - a French general officer at the time who later served as the Chief of State of Vichy France from 1940 to 1944 -  led people to collaborate with the German occupant and the Vichy regime. Uniquely in the literature, we digitized a newly declassified dataset on nearly 100,000 individuals who collaborated during World War II. We also construct a new dataset of all the location of the regiments that fought during World War I, in particular during the Battle of Verdun Pétain is consider the "victor" of. We find that communes where regiments fought the Battle of Verdun under direct command of Pétain in 1916, compared to similar communes that fought the same Battle under a different general, had 30% more collaborators during WWII. We show that rotation at Verdun under Pétain was as good as randomly assigned. Yet, in those communes, more people joined collaborationist paramilitary groups, militated in collaborationist political organisations, or directly collaborated with the Nazi regime e.g by joining the Gestapo. We also provide evidence that this translated into a durable Left-Right swing in post-War France.
(joint with Anna Dagorret, Pauline Grosjean and Saumitra Jha)

We provide a quantitative assessment of the impact of imports from low-wage countries (LWCs) on CPI inflation in France during 1994-2014, using detailed micro data on imports and exports. The share of imports from low-wage countries in consumption increased from about 2% to 7%, and resulted in a negative impact on CPI inflation of about 0.17 pp per year on average. This effect decomposes in three channels. 1) The substitution channel, capturing the replacement of domestic production by goods from LWCs, accounts for almost -0.05 pp. 2) The rise in the proportion of LWC goods in total imports weighed down on imported inflation. This channel reduced French CPI inflation by 0.06 pp per year. 3) Instrumental variable estimation of the competition channel at the product level shows that the increase in the market share of LWCs in French expenditures led to a negative effect of 0.06 pp on CPI inflation.

(joint with Erwan Gautier (Banque de France) and Sophie Guilloux-Nefussi (Banque de France))

What is the role of heterogeneous house-price expectations for boom-bust cycles in the housing market? We exploit a unique Dutch panel data set on households'house price expectations and their consumption, savings and housing choices for the period 2003-2016. This period was characterized by a pronounced boom-bust cycle in the housing market. Conditioning the sample on household heads who report non-zero house price expectations, we nd that expectations closely track realized house prices. We next develop a structural life-cycle model of the Dutch housing market where we distinguish household types according to their house price expectations. We employ a calibrated model variant to test if observed variations in expectations can account for the housing boom-bust cycle. First results show that our model closely matches the observed fluctuations of the rent-to-price ratio in the data but overshoots the size of the housing boom.

This paper studies the optimal time-consistent distortionary taxation when the repay- ment of government debt is not enforceable. The government taxes labor income or issues non-contingent debt in order to finance an exogenous stochastic stream of fiscal shocks. Debt can be repudiated subject to some default costs. Optimal policy is characterized by two opposing incentives: an incentive to postpone taxes by issuing more debt for the future (“Greed”), and an incentive to tax more currently in order to avoid punishing de- fault premia (“Fear”). A Generalized Euler Equation (GEE) captures these two effects and determines the optimal back-loading or front-loading of tax distortions. Even if de- fault risk is small, tax-smoothing is severely limited. The same mechanisms operate also in environments with long-term debt.

We introduce advertising congestion along with a time-use model of consumer choice among media. Both consumers and advertisers multi-home. Higher equilibrium advertising levels ensue on less popular media platforms because platforms treat consumer attention as a common property resource: smaller platforms internalize less the congestion from advertising and so advertise more. Platform entry raises the ad nuisance “price” to consumers and diminishes the quality of the consumption experience on all platforms. With symmetric platforms, entry still leads to higher consumer benefits. However, entry of less attractive platforms can increase ad nuisance levels so much that consumers are worse off.

The political resource curse is the idea that natural resources can lead to the deteriorationof public policies through corruption and rent-seeking by those closest to political power. Oneprominent consequence is the emergence of conflict. This paper takes this theory to the datafor the case of Mozambique, where a substantial discovery of natural gas recently took place.Focusing on the anticipation of a resource boom and the behavior of local political structuresand communities, a large-scale field experiment was designed and implemented to follow thedissemination of information about the newly-discovered resources. Two types of treatmentsprovided variation in the degree of dissemination: one with information targeting only localpolitical leaders, the other with information and deliberation activities targeting communitiesat large. A wide variety of theory-driven outcomes is measured through surveys, behavioralactivities, lab-in-the-field experiments, and georeferenced administrative data about local con-flict. Information given only to leaders increases elite capture and rent-seeking, while infor-mation and deliberation targeted at citizens increases mobilization and accountability-relatedoutcomes, and decreases violence. While the political resource curse is likely to be in play,the dissemination of information to communities at large has a countervailing effect.

Home visiting programmes are increasingly being implemented for expectant families and those with young children identified with vulnerabilities or risks for poorer outcomes, however evidence on their long-term impacts is surprisingly scarce. In this paper we study the health impacts at the 18-year follow-up of the Memphis randomized controlled trial of the Nurse Family Partnership (NFP), one of the oldest and most thoroughly evaluated nurse home visitation programs in the United States. We find lower rates of hypertension in the mothers who were visited by the nurses in pregnancy and postnatally, and lower prevalence of obesity in their adolescent children. These effects are mostly concentrated in girls and their mothers, and can be traced back since the prenatal and the birth period. Our results provide importance new evidence on the potential role of prenatal and infancy home visiting in preventing or delaying the onset of costly chronic health conditions.

Classic and modern theories of rebel warfare emphasize the role of unexpected attacks against better equipped government forces. We test implications of a simple model of combat and information-gathering using highly detailed data about Afghan rebel attacks, insurgent-led spy networks, and counterinsurgent operations. Timing of rebel operations responds to changes in the group's access to resources. Results are supplemented with numerous robustness checks as well as a novel IV approach that uses machine learning and high frequency data on local agronomic inputs. Main effects are significantly enhanced in areas where rebels have the capacity to spy on and infiltrate military installations. Consistent with the model, shocks to labor scarcity and government surveillance operations have the opposite effect on attack timing.

This paper tests whether worryingabout rainfall risk impairsfarmers’ decision-making, through worse cognition. Behavioraltheories predict that worries could impose a psychological tax on farmers, reducing the quality of their decisionsand, in turn,leading to material consequences at all times and across all states of nature –even when negative rainfall shocks do not materialize down the line. Using a novel technology to run lab experiments in the field, we combine survey experiments and recent rainfall shocks to estimate the effects of worryingabout rainfall risk on farmers’ cognition.We find that worries about rainfall increase farmers’ cognitive load and their susceptibility to a variety ofbehavioral biases. In theory, insurance could mitigatethose effects by alleviating the material consequences of rainfall risk. To test this hypothesis, we randomly assignoffers of an index insurance product, and find that it does not affect farmers’ worries about rainfall, cognitive load,or susceptibility to biases. These resultssuggest that farmers’ anxiety might berelatively difficult to alleviate.

 

This paper uses administrative tax data to examine the long-term effects of an experimental job-search assistance program operating in Nevada in 2009. The program required randomly-selected unemployed workers who had just started collecting unemployment insurance (UI) benefits to undergo an eligibility review and receive personalized job-counseling services. The program led to substantial short-term reductions in UI receipt, and to persistent, long-term increases in employment and earnings. The program also affected participants’ family outcomes, including total income, tax filing, tax liability, and home ownership. These findings show that job-search assistance programs may produce substantial long-term effects for participants and their families.
(with M. Michaelides and A. Patel)